Memorandum submitted by Cédric
Philibert, International Energy Agency
1. I am employed by the International Energy
Agency, but I want to make clear that I am not speaking on behalf
of the IEA or its member governments. No implication of agreement
by the IEA or its 27 member States with the views I express here
should be made.
2. Over the last 15 years the International
Energy Agency has conducted extensive research on many aspects
of climate change and climate change mitigation, providing statistics
and projections of current and future energy-related CO2
emissions, offering perspectives on climate-friendly energy technologies,
developing recommendations for energy efficiency policies, and
analysing policy tools for greenhouse gas reductions from both
theoretical and practical perspectives.
3. Over the last 10 years the Energy Efficiency
and Environment Division at the IEA has also assessed various
options for further engaging developing countries in the post-2012
climate mitigation effort, and facilitating the elaboration of
a global mitigation framework that should include as many countries
as possible, starting with all major emitters.
4. The concentration level and the timing
of emission reductions in the ultimate objective of the Convention
have not been spelled out. As the IPCC says, "determining
what constitutes "dangerous anthropogenic interference with
the climate system" in relation to Article 2 of the UNFCCC
involves value judgments", but "science can support
informed decisions on this issue." If long term temperature
change is to be limited to any level below 3 degrees C, global
emissions should return to 2000 levels by 2050, and preferably
lower. If long term temperature change is to be limited to 2 degrees
C, CO2 emissions would need to peak before 2015 and
then decrease to 15% to 50% of 2000 levels, by 2050.
5. For these two scenarios to remain achievable,
reducing or even eliminating greenhouse gas (GHG) emissions of
developed countries is not enough. Under the most likely scenarios,
the unabated emissions of developing countries would alone exceed
global levels compatible with these stabilization scenarios. For
example, energy-related CO2 emissions alone, from developing
countries only, would reach 32 Gt CO2 by 2050 under
the Baseline Scenario of our Energy Technology Perspectives
2006—that is 12.5% higher than global emissions in 2000.
The world would most certainly exceed the 2 or 3 degrees C temperature
change mentioned above.
6. Emissions are growing most rapidly in
developing countries. China is now the largest emitter of CO2
from fossil fuel use, with the USA as second. India will soon
rank third. Taking in account all greenhouse gases and sources,
including agriculture and deforestation, India, Indonesia and
Brazil are among the fifth largest emitters.
7. Developing country participation in a
future framework to reduce GHG matters for other reasons than
their emissions growth rate. Some of our energy-intensive industries
face constraints that their competitors in developing countries
do not. It is politically difficult to envision a framework that
could, in the end, encourage re-location to the developing world,
at the expense of our industrial output. Such re-location could
also hamper the environmental goal if it leads to a relative increase
in emissions, or leakage. Finally, the United States is unlikely
to accept a significant effort to cut emissions if major developing
countries do not take part as well.
8. I will consider the need for various
options for future action by developing countries, and what these
options might be. I will then suggest a possible progression from
the softer and narrower ones to the harder and broader ones. I
will finally offer some concluding thoughts.
THE NEED
FOR NEW
OPTIONS
9. Developing countries have adamantly refused
to take on binding commitments on their emissions during the negotiations
that led to the Kyoto agreement. In most developing countries,
per capita emissions and, more importantly, per capita
income, are significantly lower than those of most industrialised
countries. Hence the distinction between developed countries—and
among them countries in transition—and developing countries
among Parties in the Kyoto Protocol, based on the Climate Convention's
principle of "common but differentiated responsibilities
and respective capabilities". While all countries have accepted
the commitment to provide emission inventories and to adopt policies
and measures to mitigate emissions, only developed countries have
accepted binding commitments relative to their GHG emissions.
10. These commitments have themselves been
differentiated amongst developed countries. Further, within the
European Union itself, the burden-sharing agreement was an essential
ingredient of the solidarity and unity amongst all Member States.
11. Future mitigation action by developing
countries as part of a future mitigation architecture could require
further differentiation in addition to the emission target. The
nature of their commitments could also be diversified, to fit
their national circumstances, including their economic and emissions
evolution since 1992, when the distinction between developed and
developing countries was cast in the Climate Convention. The notion
that binding emission caps may represent a significant constraint
on economic development is still strong in developing countries,
and cannot be dismissed easily, given the many uncertainties on
unabated emission trends and emission reduction potential. Alternative
options must be explored, including those that may be transitory.
This is why, since 2000, we have identified five broad options:
12. These options are still today the most-frequently
mentioned, although sometimes under different names. Furthermore,
over the years the issues of adaptation, financing and technology
transfer have progressively been given a higher profile in the
negotiations, alongside mitigation. Nevertheless, mitigation will
remain the focus of this testimony, while obvious links with financing
and technology transfers will be outlined.
13. The Bali Action Plan, adopted at COP-13,
calls for the consideration of "nationally appropriate mitigation
actions by developing country Parties in the context of sustainable
development, supported and enabled by technology, financing and
capacity-building, in a measurable, reportable and verifiable
manner".
14. My own view is that, wherever appropriate,
the various options that are compatible with quantitative objectives
and flexible mechanisms such as those in the Kyoto Protocol can
be tailored to satisfy, at least in part, the requirements of
the Bali Action Plan. For example, flexibility mechanisms directly
reward emission reductions below an agreed target, in a measurable,
reportable and verifiable manner.
POLICIES AND
MEASURES
15. All Parties under the UNFCCC are supposed
to introduce policies and measures to curb their greenhouse gas
emissions. Current trends suggest that this element of the Convention
is far from effective. Some policies and measures could conceivably
be made "mandatory" in a future agreement. Having taken
part to the negotiations of the Berlin Mandate in 1995 I can testify
that, at that time at least, the developing countries were as
strongly opposed to this possibility as they were to binding emission
targets.
16. Another option would be to ask developing
countries to commit themselves to various policies and measures.
For some analysts, these policies should primarily aim at meeting
the (sustainable) development needs, and curb the growth in GHG
emissions as a secondary benefit—hence the often-quoted
label "SD-PAMs" (for Sustainable Development Policies
and Measures).
17. Commitments to implement specific policies
and measures that lead to lower emissions could thus be made binding,
with some commitment by developed countries to assist developed
countries in the process of implementation. As an example, there
are large possibilities for reducing emissions from energy efficiency
improvements that would be self-rewarding due to energy savings,
but may require help from more advanced nations.
18. Another possibility is to reward such
policies through some new form of "carbon credits",
which developing countries could sell on the international carbon
market. However, "crediting" policies and measures is
analytically complex because of difficulties in estimating the
effects of such policies. Furthermore, crediting emission reduction
policies will encourage developing countries to select "good
policies", leaving out those that lead to rising emission
levels. A better option might be to leave the countries committed
to various policies with the option to seek for further reward
through the flexible mechanisms with sector-wide or country-wide
non-binding targets.
19. Commitments to some policies and measures
could provide a useful starting point for developing countries
to collect and document relevant policies in a systematic and
harmonised manner—and offer an opportunity for these types
of actions to be officially recognised.
SECTORAL TARGETS
20. Sectoral targets could be of different
types, but I focus here on country-specific quantitative sectoral
targets. In this model, a country's initiative limited to a sector
is recognised by the international community (eg UNFCCC Parties).
Here, one could envision the possibility to credit greenhouse
gas emission reductions on a sectoral basis. Sectoral targets
of this kind could be binding or not.
21. A strong argument in favour of sectoral
targets may come from the recognition that when it comes to energy/CO2
performance, the frontier between developed and developing countries
becomes blurred. While on average OECD countries' industry tends
to use energy more efficiently, the most efficient plants are
sometimes found in developing countries. If capacity and technology
are available in these regions, the central question is how to
promote their broader diffusion. Another possible advantage of
sectoral targets is to allow a focus on the most advanced sectors
in a developing country, and those that operate in the same markets
and face the same price as their developed country counterparts.
22. There are various possibilities for
setting sector-wide targets. Companies usually favour benchmarking,
objectives set per unit of output. The data needs are not trivial,
and the definition of output may be difficult in some sectors—not
to mention the monitoring aspects of such policy instruments,
in countries where government capacity is lacking.
23. Interestingly, the proposal for the
revision of the EU emissions trading scheme refers to such sectoral
approaches, in its discussion of trade-exposed industries that
may suffer from emissions leakage if outside competitors do not
bear the cost of CO2 emissions. Sectors possibly at
stake include cement, iron and steel, aluminium, paper and pulp,
glass, oil refining, to name a few. This option is drawing a lot
of attention, although no obvious solution exists to bring together
actors facing hard emission constraints in our countries, with
those enjoying a free ride in the rest of the world. I would note
that the risk of trade measures is never far.
24. If sectoral targets is the main driver
of emission reductions in the developing world, the energy supply
and energy consuming sectors (heat and power, buildings, industry,
transport), but also, depending on the country, the agriculture
and forestry sectors, could be considered more broadly. One obvious
candidate may be power generation, the largest and most rapidly
growing source of CO2 in the developing world. Emissions
and output are relatively easy to monitor and there is good experience
in setting baselines, thanks to the Clean Development Mechanism.
NON-BINDING
TARGETS
25. A country subject to a non-binding target
would be allowed to sell emission allowances for any reduction
below its target—but not forced to buy allowances if emissions
were higher. No constraint on emissions could thus risk constrain
the economic development. This is why they are often named "no-lose
targets".
26. Various levels for non-binding targets
can be identified. A first level would be that of future unabated
emissions, following business-as-usual trends. All emission reductions
would be rewarded through the flexible mechanisms, ie at the expense
of other countries. A second, lower level would be that of future
emissions when "win-win" reduction possibilities—energy
savings, ancillary benefits—are taken in account. Only emission
reductions below that level would be made at the expense of other
countries.
27. The non-binding nature of the target
may provide a different negotiating atmosphere. For it is non-binding
it cannot be perceived as a threat for economic development. Developing
countries will be negotiating the size of an advantage, not the
level of a constraint. The perspective of their partners in the
negotiating process would also evolve. Stringent targets always
look better. But too stringent a target increases the risk that
it ends up overtaken by actual emissions and ignored. There is
thus a kind of common interest of all negotiators in negotiating
in good faith a realistic, achievable target.
INDEXED TARGETS
28. Indexation would allow for revising
assigned amounts as economic growth deviates from shared expectations—with
a great variety of possible rules and importance of revisions.
It would be wrong to state that indexed targets are weak by their
sheer nature. A fixed target can be set at high levels and bring
hot air. Indexed targets can be exactly as ambitious as fixed
targets—or more, as they remove part of the cost uncertainty.
29. "Intensity" targets, where
the targets are expressed in emissions per GDP unit, represent
an extreme form of indexed target. I would favour partial indexation
only. If growth was more rapid than expected it also drove more
rapid rotation of capital stock and provided more opportunities
for abatement. If growth was less rapid than expected, reducing
allowed emissions may become a double pain in sluggish economies,
since some emissions are relatively independent from economic
activities.
BINDING TARGETS
30. I consider here "fixed" binding
targets like those currently taken by Annex I countries in the
Kyoto Protocol. Parameters that might be considered in this option
include country ranking, timing, thresholds, and stringency of
commitments. These parameters might be developed individually,
or in the framework of burden sharing of a specified global emission
or concentration target. We will also briefly discuss here the
possible introduction of price caps in the international commitments.
31. Procedures have been suggested for negotiating
legally binding absolute targets for non-Annex-I countries. They
usually recognise that these targets would be "growth targets":
a developing country commitment would likely be set at some level
above the country's current emission level. This was already the
case of some Annex-I countries in the Kyoto Protocol, and some
others through the European joint-fulfilment agreement, even though
industrialised countries' emissions are, on aggregate, capped
at 5.2% below their current levels.
32. An important subset of proposals tries
to draw countries' commitments from a global objective for emissions,
concentrations or even temperature change, largely in an effort
to promote equity. This is notably the case for the "Contraction
and Convergence" scheme. Other allocation approaches start
also from the need to achieve convergence but distinguish among
the various sectors of the economies. Under the "Global Triptych
Approach", for example, one would base allocation on the
convergence of greenhouse gas intensity for the power generation;
on the convergence in energy efficiency for the energy intensive
industry; and on per capita GHG emissions for the domestic
sector that includes transportation and the residential/small
business sector. Other analyses have extended such approaches
to more sectors and more gases.
33. Developing countries accepting fixed
and binding targets will recall their legitimate need to develop
their economy and likely seek for an allocation that binds but
does not bite—ie over and beyond the highest projections
of future unabated emissions. Uncertainty in unabated trends and
technology developments, is at stake here. While industrialised
countries may still benefit from extending the mitigation framework
to developing countries through accessing large potentials of
presumed cheap emission reductions, the obligation of making large
payments to other countries—not only for actual reductions
but also for hot air trading—can raise political difficulties.
34. One possible way for alleviating the
risk that binding targets entail unexpectedly high costs is to
introduce price caps ("safety valves"), ie the possibility
for a country (or emission sources within a country) to acquire
additional allowances at a set price. Indeed, this possibility
could be open to industrialised countries as well. Provided the
price level is set sufficiently high, ie in the upper range of
cost expectations resulting from a given target, the option may
help countries adopt targets relatively more ambitious than in
its absence. It may prove an essential ingredient for bringing
some countries, industrialised or developing, into the international
mitigation architecture. For climate change is driven by the accumulation
of greenhouse gas in the atmosphere, the precision in achieving
a given objective in any specified short period of time may be
less important than the overall ambition of the mitigation scheme—as
well as the breadth of its coverage.
35. Developing countries accepting binding
targets could conceivably allow their economies some protection
against possible high carbon prices with price caps that could
be lower, although the level of efforts will likely be differentiated
in the allocation process. Trading amongst areas with uneven price
cap levels remains possible—one simply needs to make sure
that a country ends up a net seller only if it is in full compliance
ie does not "use" its price cap. A prerequisite for
this option, though, might be that industrialised countries too
face financial consequences if their emissions exceed their target,
whether these consequences take the form of financial penalties
or of price caps (with no obligation to restore their target)
POSSIBLE GRADUATION
36. The future GHG mitigation framework
could incorporate various types of quantitative objectives. I
would like to suggest a possible progression. UN agencies often
distinguish "low-income", "middle-income",
and "high-income" developing countries.
37. In the large group of low-income countries,
with per capita GDP below USD 1,000, they further distinguish
the "least-developed countries", characterised by a
"human resource weakness criterion"—and a relatively
small size. These 50 countries currently total about 750 million
inhabitants. Commitments on various policies and measures, with
the help of the rest of the world, might be the more realistic
option for these least-developed countries.
38. There are 15 low-income developing countries
that are not LDCs. Seven—China, India, Indonesia, Pakistan,
Nigeria, Philippines and Vietnam—total more than three billion
inhabitants. The emissions from some sectors of three of them
would rank higher by size than the overall emissions of various
other countries. Heat and power generation in China and India,
Industry in China and India, Agriculture in China all emit each
year more than half a billion tonnes CO2-Eq. Sectoral
targets and/or non-binding targets could offer realistic possibilities
to these countries.
39. Indexed targets could be an option for the
50 "middle-income" developing countries, with GDP per
capita below USD 4,500, which is close to the per capita
income of the less-wealthy Annex-I countries by 1997 at the time
of the Kyoto Protocol. Brazil, Egypt, Iran, Thailand and South
Africa are the most important ones. Note that Agriculture and
Land-Use Change in Brazil also emit each year more than half a
billion tonnes CO2-Eq each.
40. Binding targets—not excluding
the possible inclusion of safety valves—could be an option
for the 50 "high-income developing countries", whose
per capita GDP is today higher than the level of wealth
reached in 1997 by the less-wealthy industrialised countries having
accepted such commitment in the Kyoto Protocol. The largest are
Mexico, Korea, Argentina, Saudi Arabia and Venezuela.
41. I must underline a few aspects of this
suggestion. First, though it does include some elements of human
resources and size (for LDCs), it is mostly based on per capita
income levels. This I believe is more relevant than per capita
energy consumption or per capita CO2 or CO2-equivalent
emissions, which are too much dependant on a wide range of national
circumstances—even more so than total CO2 or
GHG emissions, which primarily reflect the size of a country,
only one possible dimension of its capacities.
42. If the forthcoming framework is based
on some flexible mechanisms as one may wish, then what matters
first is the level of wealth—the willingness to pay for
climate stabilisation, the financial and technological capabilities,
the purchasing power of the various populations. To some extent
this indicator also captures elements of historical responsibility
and capability.
43. The grouping of countries mentioned
here is that of the UN system, and is based on exchange rates.
Purchasing power parities (PPP) would elevate many developing
countries, including China and even India, to the level of wealth
of the less-rich industrialised countries by 1997. This does not
seem entirely realistic to me, in particular as mitigating climate
change will require the development and transfer of internationally
tradable lean carbon emitting technologies. PPP do not adequately
reflect the capacity of a country to be present on international
markets, as exchange rates do. However, as this is only one dimension
of climate mitigation, one may seek for more complex indicators
mixing these two measures of income.
44. Let me also outline the very rough nature
of the progression I have suggested here. Whatever the level of
development of a given country, a non-binding target set at a
relatively low level might be preferred to a binding one set a
too high a level. Good policies might be preferred to too generous
binding sector targets. In a nutshell, no option can be deemed
superior until numbers have been agreed, and details worked out.
CONCLUDING THOUGHTS
45. First, on the issue of technology development
and transfer, and financing. If the international mitigation framework
is based on some form of differentiated quantitative emission
targets and flexible mechanisms, financing and technology development
and transfer will come along. This does not mean that other forms
of international cooperation are not useful—we at the IEA
host about 40 implementing agreements about energy technology,
from end-uses to renewable to carbon capture and storage, and
we welcome developing country participation.
46. This does not mean either that new specific
mechanisms could not help, in particular in the framework of policies
and measures. One example of particular relevance would be the
energy efficiency partnership currently under discussions. It
means that climate negotiators should always remind or be reminded
that accepting some form of an emission targets in itself may
bring its counterpart in financial and technological terms.
47. Second, on the Clean Development Mechanism.
It suffers from various limitations. Nuclear power has been excluded,
carbon dioxide capture and storage has not—not yet?—been
included. Halting deforestation is not in, reforestation projects
are limited. Energy efficiency projects have difficulties finding
their way in the additionality assessment procedures. Low demand
and low prices do not favour renewable energy projects either.
While the CDM would remain a useful tool in countries with no
emission cap of any kind, its greater merit may be to have paved
the way for broader mechanisms.
48. Third, on the issue of reducing emissions
from deforestation in developing countries (REDD). Except for
possible methodological uncertainties I would see no specific
reason to exclude emissions or removals from land-use change and
forestry from a country-wide target of a developing economy, as
is the case for industrialised countries in the Kyoto Protocol.
49. Finally, I would like to draw your attention
to some specificities of energy use in developing countries. About
2.4 billion people in developing countries depend on biomass for
cooking and (water) heating. Collecting this biomass is time-consuming
and often not sustainable. Its combustion is often inefficient,
incomplete and polluting—with large health damage from in-door
air pollution. While biomass use can be made more effective and
healthier, its substitution with cleaner fossil fuels, such as
kerosene and, preferably, LPG, is often an easy way to rapidly
improve the living standard of the poor. Negotiators must take
care that country-wide GHG targets in developing countries, binding
or not, do not result in halting the transition from dangerous
biomass uses to cleaner fossil fuel use, whose aggregate impact
on GHG emissions will likely remain marginal.
January 2008
| 
